Development and Validation of the Praise, Indulgence, and Status Parenting Scale

Abstract

The overarching aim of this study was to develop and validate a new scale (i.e., the Praise, Indulgence, and Status Parenting Scale [PISPS]) to measure modern parenting practices and behaviors consistent with instilling ideals of specialness (i.e., the notion that one is special), self-esteem, and status in children. In 2 studies on emerging adults (Study 1: N = 582, M_age = 19.46; Study 2: N = 464, M_age = 19.58), the PISPS was developed and validated using classical test theory (Study 1) and further refined using item–response theory (Study 2). Results from both studies indicated a 3-factor structure with factors differentially linked with correlates of interest including parenting strategies, self-esteem, narcissism, entitlement, and internalizing symptoms. Study 3 further validated the PISPS in a sample of parents (N = 638, M_age = 35.79) reporting on their parenting and their child’s emotion regulation and symptoms of psychopathology. Overall, findings support the PISPS, its psychometric properties, and its unique contribution to child symptoms.

The desire to facilitate children’s healthy development and well-being is arguably one of the more central goals of society. For decades, the role of parenting in child development has received considerable attention, yielding important theoretical and empirical models of parenting and parent–child relationships (e.g., Baumrind, 1989; Gottman, Katz, & Hooven, 1996). Research in the areas of child development and well-being has highlighted a host of beneficial parenting practices that foster healthy development, an approach Holden, Ashraf, Brannan, and Baker (2016) described as positive parenting. Coinciding with this parenting lens has been a focus on children’s self-esteem—or the “the level of global regard that one has to the self as a person” (Harter, 1993, p. 88)—as low self-esteem has been associated with poor outcomes such as depression, poor health, aggression, and antisocial behavior, even when controlling for potential covariates such as gender, socioeconomic status, and potential reporter bias (e.g., Donnellan, Trzesniewski, Robins, Moffitt, & Caspi, 2005; Sowislo & Orth, 2013; Trzesniewski et al., 2006). Investigators have highlighted the complex interplay between healthy or adaptive self-esteem and inflated or unstable self-esteem as well as other deleterious outcomes such as narcissistic traits, entitlement, and psychopathology (e.g., Brummelman, Thomaes, & Sedikides, 2016; Trumpeter, Watson, O’Leary, & Weathington, 2008). In this study, we developed and validated a parent self-report questionnaire to better understand parenting practices that would contribute to this continuum of outcomes. This scale has the potential to offer a more nuanced approach to understanding parenting that relates to adaptive self-esteem, inflated self-esteem, narcissism, entitlement, and ultimately, well-being.

Theoretical framework: Symbolic interactionism

Symbolic interactionism posits that children develop a sense of self by internalizing symbolic representations arising from significant relationships with others (e.g., the parent–child relationship; Harter, 1999). From this perspective, parents’ views of their children and the ways in which these views are communicated are critical to child development. For instance, parenting approaches that offer warmth and praise provide a secure and supportive environment for children, which in turn allows them to develop adaptive self-esteem and noncontingent self-worth. In contrast, research has shown that parental overvaluation—that is, “parents’ belief that their own child is more special and more entitled than other children” (Brummelman, Thomaes, Nelemans, Orobio de Castro, & Bushman, 2015, p. 666)—indulgence, and a focus on the pursuit of extrinsic goals is associated with narcissism and entitlement. Such messages might convey to children that they are superior to others and more deserving than others. A focus on the pursuit of extrinsic goals in particular communicates to children that their self-worth is contingent on successes in those arenas and might predispose them to anxiety and depression.

As mentioned previously, the interplay between healthy or adaptive self-esteem and inflated or unstable self-esteem as well as other deleterious outcomes such as narcissistic traits, entitlement, and psychopathology, is likely complex. Although these outcomes might be related, investigators have documented that they are indeed distinct constructs (Brummelman et al., 2016; Trumpeter et al., 2008). For example, Brummelman et al. (2016) discussed the core distinction between narcissism and self-esteem as being in the “perception of regard from others” (p. 8) whereby narcissism is associated with the perception that one is superior to others and self-esteem is associated with the perception of one’s sense of adequacy (i.e., self-worth). In line with this perspective, we posit that distinct parenting practices are associated with each of these outcomes.

A proposed spectrum of parenting practices

Consistent with the aforementioned theoretical framework, relevant parenting approaches likely also follow a spectrum of distinct parenting practices ranging from praising one’s child to overvaluing or overpraising and encouraging children to pursue extrinsic goals. Consistent with Trumpeter et al.’s (2008) framework, we were interested in the spectrum of outcomes ranging from adaptive self-esteem to narcissism and psychological symptoms. We were also interested in entitlement and emotion dysregulation as key outcomes given their correlations with narcissism and psychopathology, respectively (e.g., Di Paula & Campbell, 2002). Here, we briefly review relevant empirical findings in the parenting literature as they relate to each of these outcomes.

Parenting has been associated with self-esteem in both cross-sectional and longitudinal research, although empirical findings are largely mixed in longitudinal research predicting self-esteem in adolescence (Harris et al., 2015). Despite this, empirical and theoretical findings are consistent in terms of the specific parenting practices that are positively associated with self-esteem, namely parental warmth and support (Murray, Holmes, & Griffin, 2000). In contrast, parental rejection, indulgence, and overcontrol are associated with low self-esteem. It is theorized that parental indulgence and overcontrol impede children’s opportunities to gain and internalize accomplishments. Rather, children whose parents are overinvolved might attribute successes to those involved figures (i.e., parents as in the case of parental overinvolvement; DeHart, Pelham, & Tennen, 2006).

Previous theoretical and empirical research has also demonstrated important links between parenting and narcissism (Claudio, 2016; Kernberg, 1975; Millon, 1981), with likely variation dependent on how narcissism is defined. Narcissism is a constellation of personality traits that include an inflated sense of self-worth, a lack of concern or empathy for others, entitlement, and seeking admiration of others (American Psychiatric Association, 2013). More recent research identifies two mostly distinct dimensions of narcissism: grandiose narcissism (e.g., grandiosity, callousness, manipulativeness, exhibitionism, and a domineering interpersonal approach) and vulnerable narcissism (e.g., self-absorption, distrust, hostility, psychological distress, insecurity, and fragility; Miller & Campbell, 2008). Grandiose narcissistic traits (as measured via the Narcissistic Personality Inventory; Raskin & Terry, 1988) have been linked to several parental constructs such as warmth and psychological control (Horton, Bleau, & Drwecki, 2006), permissiveness (Watson, Little, & Biderman, 1992), indulgent parenting (Cramer, 2011), and parental overvaluation of children, although these links tend to be rather modest in size. Research suggests vulnerable narcissism might relate differently with parenting such that it is generally related to low parental warmth, low monitoring, and high intrusiveness (Miller et al., 2010). Taken together, these findings suggest that grandiose narcissism is related to parenting that is overly warm and overvaluing, whereas vulnerable narcissism is related to low parental warmth and intrusiveness or overcontrol.

Entitlement is a central correlate of narcissism, and as such, is particularly relevant to the current scale development. Theory suggests that parental socialization (e.g., parental overvaluation) is related to the development of entitlement in childhood (Thomaes & Brummelman, 2018), and empirical research supports this tenet. For example, research has shown that indulgent parenting is associated with increased entitlement, stress, and depressive symptoms in young adults (e.g., Barton & Hirsch, 2016). Goal contents theory (a minitheory of self-determination theory; Ryan & Deci, 2002) states that the pursuit of extrinsic goals (vs. intrinsic goals) is negatively associated with well-being. Indeed, research supports positive links between parenting that is overinvolved and focused on achievement and extrinsic goals and maladaptive outcomes (Givertz & Segrin, 2014; Soenens, Wuyts, Vansteenkiste, Mageau, & Brenning, 2015). Taken together, research suggests that parental overvaluation and encouragement for children to seek extrinsic goals contribute to entitlement in young people.

Parental praise, support, and encouragement foster child autonomy and self-reliance, which is key to child well-being. In contrast, practices that not only convey specialness but also overvalue the child’s characteristics are likely associated with poor self-esteem and high levels of narcissistic traits. Consistent with Adler’s self-reliance theory (Adler, 1930), overindulgent parenting prevents children from developing adaptive self-reliance required for well-being. Repeated messages that convey to children that they are the best and most special might, over time, negatively affect their ability to manage challenging experiences (i.e., regulation), including emotional experiences, which would in turn affect their mental health. This would be especially relevant when a child’s performance is inconsistent with parental messages he or she received while growing up. The child might then be left with few resources and supports to tackle the challenging situation.

Taken together, we were interested in three primary parental practices that could be associated with adaptive self-esteem, narcissism, entitlement, emotion dysregulation, and psychological symptoms: praise, indulgence, and status. In this study, praise was conceptualized as parental messages that the child is special, unique, talented, and worthy. In contrast, indulgence was defined as parental practices and messages that overpraise and overvalue the child, which might also be communicated as superiority over others. Finally, given important links between the pursuit of extrinsic goals and negative outcomes, we included parenting practices that encourage children to seek status and fame.

This study

To measure the spectrum of distinct parenting practices described previously, we developed and validated the Praise, Indulgence, and Status Parenting Scale (PISPS). Over three samples, we developed and then refined the items reflecting these parenting practices. In Study 1, we used classical test theory to develop an initial item set, which we then validated in a sample of young adults retrospectively reporting on their mothers’ parenting practices. In Study 2, we refined the scale using item–response theory (IRT) analyses and given that mothers and fathers might differ on these parenting practices (e.g., Cowan, Cowan, & Kerig, 1993), we expanded the initial scale to include a father version. We investigated the nomological network of the scale in both studies by examining links to other well-established parenting constructs (i.e., warmth, rejection, overcontrol) as well as key outcomes: self-esteem, narcissism (grandiose and vulnerable), entitlement, and internalizing problems (in Study 2 only). These specific parenting constructs were chosen based on their relevance to the correlates being examined. Finally, in Study 3 we extended the PISPS to a sample of parents of middle childhood children and examined relations to child emotion regulation and symptoms of psychopathology. We chose this developmental period based on previous research documenting that self-esteem and narcissistic traits tend to emerge early in middle childhood (Thomaes, Stegge, Bushman, Olthof, & Denissen, 2008). For all studies, we examined the unique contribution of the PISPS above and beyond previously established parenting constructs.

In terms of specific hypotheses, we expected PISPS praise to be positively associated with parental warmth and negatively associated with parental rejection. Indulgence and status were expected to relate to parental overprotection and overvaluation. Regarding outcomes, we expected parental praise to be associated with positive outcomes such as enhanced self-esteem and emotion regulation, but not with negative outcomes such as narcissism, entitlement, and psychological symptoms. We expected indulgent parenting to be associated with negative correlates such as low self-esteem, high levels of narcissistic traits, entitlement, emotion dysregulation, and psychological symptoms. Finally, we expected status to be associated with low self-esteem, high levels of narcissistic traits, entitlement, emotion dysregulation, and psychological symptoms. We also expected the PISPS to account for significant variance in outcomes, above and beyond established parenting constructs such as warmth, rejection, overprotection, and overvaluation.

Study 1

Method

Praise, Indulgence, and Status Parenting Scale development

The development of the PISPS followed the criteria set forth by Clark and Watson (1995). Items were generated in two phases. The first phase included a review of the relevant literature to inform the development of scale items of caregiver behaviors consistent with ideals of child specialness, self-esteem, and achievement. Initial items were generated by the first and second authors. Phase 2 of item generation entailed eliciting caregiver behaviors from four experts in parenting, personality, narcissism, and externalizing behavior research at the University of Georgia. An evolving and iterative process occurred such that experts incorporated items and provided feedback on items generated from others over multiple discussions. The final list included 35 caregiver behaviors (see Table 1).

Table 1.

Factor loadings from Study 1.

Items	Factors
	Praise	Indulgence	Status
1. My mother told me I was special	0.80	−0.08	0.02
2. My mother thought it was funny when I broke the rules	0.02	0.56	0.17
3. I never got away with anything with my mother	Dropped
4. My mother would call me pet names such as “Little Princess” or “Handsome Little Prince”	0.54	0.14	0.09
5. My mother told me I was unique	Dropped
6. My mother told me I could be anything I wanted	0.75	−0.23	0.04
7. My mother told me I was the most beautiful, handsome, or pretty child	0.80	0.04	0.01
8. My mother told me that I was smarter or brighter than other kids	0.64	0.11	0.11
9. My mother made me feel like I was the most special kid around	Dropped
10. I received special (better) treatment than my peers	0.51	0.30	0.16
11. My mother told me I was going to do great things	Dropped
12. My mother told me I would be successful no matter what	0.73	0.01	0.18
13. My mother told me that my teachers did not understand how special I was	0.19	0.52	0.38
14. My mother made sure I got special privileges at school	0.10	0.58	0.43
15. My mother gave me most everything I wanted	0.34	0.41	0.02
16. My mother told me I did not have to do chores or other tasks because I was too precious	0.02	0.81	0.32
17. My mother told me I was a gift	0.70	0.14	0.21
18. My mother let me rule the house	0.02	0.83	0.18
19. My mother did not allow me to do anything before chores	Dropped
20. My mother told me that I was a “Godsend”	0.43	0.30	0.26
21. My mother intervened when others criticized me	0.20	0.35	0.20
22. My mother tolerated my temper tantrums	0.01	0.67	0.06
23. My mother gave in when I came home late without explanation	Dropped
24. My mother made my siblings and other children let me have first dibs on things	Dropped
25. My mother let me sit in the front seat of the car even when someone older was in the car	Dropped
26. My mother sometimes put me on a pedestal	0.28	0.57	0.20
27. My mother believed I had exceptional talents and abilities	0.63	0.07	0.20
28. My mother praised me for virtually everything I did	0.52	0.41	0.18
29. My mother rarely criticized me	Dropped
30. My mother let me make important family decisions at a young age	0.20	0.51	0.23
31. My mother told me that I had unique talents and abilities that others did not have	0.47	0.26	0.39
32. My mother encouraged me to pursue stardom	0.22	0.24	0.69
33. My mother encouraged me to be in beauty pageants or other competitions	0.09	0.32	0.65
34. My mother encouraged me to be the leader in all activities that I was involved in	0.41	0.08	0.39
35. My mother often told me, “You are going to be famous”	0.26	0.32	0.72
Internal consistency (α = .91 total scale)	α = .88	α = .83	α = .73

Note. Largest factor loading in bold.

Participants

Participants were 582 undergraduate students from a large university in the southeastern United States, who participated in the study in exchange for credit as part of a course. Participants were between the ages of 18 and 42 (M_age = 19.46 years, SD_age = 1.71, 74% female), and mostly identified as White (71%). When describing maternal caregivers, most participants reported on their biological mothers (99%), who were most likely to be married (85%). Participants reported a household yearly income ranging from less than $20,000 (2%) to greater than $80,000 per year (51%).

Procedure

Participants accessed the study online via Qualtrics where they provided demographic information and completed the PISPS about their primary female caregiver. All study procedures were in accordance with the university’s institutional review board. Informed consent was obtained from all participants.

Measures

Praise, Indulgence, and Status Parenting Scale development.

Items were presented to participants in a retrospective format where participants were asked to rate (using a 7-point Likert scale) the extent to which their caregiver engaged in the listed behaviors when the student was “young and growing up.”

Parenting.

Parenting correlates were examined via the Egna Minnen Betraffande Uppfostran–Child version (EMBU–C; Perris, Jacobsson, Linndström, Knorring, & Perris, 1980). This 23-item scale assessed parental warmth, rejection, and overprotection (e.g., “My mother showed with words and gestures that she liked me” and “My mother criticized me and told me how lazy and useless I was in front of others”). Coefficient alphas were acceptable for the three subscales (α = .81, .86, and .81 for the warmth, rejection, and overprotection subscales, respectively).

Narcissistic traits.

Narcissistic traits were assessed using the Narcissistic Personality Inventory (NPI; Raskin & Terry, 1988) and the Pathological Narcissism Inventory (PNI; Pincus et al., 2009). The NPI is a 40-item self-report measure of trait narcissism. The total score was used in this study (α = .84). The PNI is a 52-item self-report measure of grandiose and vulnerable narcissistic traits. For this study, alpha coefficients were .85 and .94 for the grandiose and vulnerable subscales.

Entitlement.

Entitlement was assessed using the Psychological Entitlement Scale (PES; Campbell, Bonacci, Shelton, Exline, & Bushman, 2004), a 9-item self-report measuring the extent to which individuals believe they deserve or are entitled to more than others. Internal consistency was good (α =.80).

Self-esteem.

Self-esteem was measured using the Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965), which is a 10-item scale designed to assess an individual’s explicit perception of self-worth. In this sample, the coefficient alpha was .84.

Data analytic plan

In line with classical test theory (Clark & Watson, 1995), the following steps were taken: (1) corrected item–total correlations were examined, and items with low correlations (i.e., r < .30) were dropped from the item pool; (2) interitem correlations were examined, and items with interitem correlations that were consistently either too low (i.e., r < .20) or too high (i.e., r > .65) were dropped from the item pool. This allowed us to obtain an item pool with interitem correlations of moderate magnitude (Clark & Watson, 1995). Subsequently, to examine the underlying structure of the items, an ordinal exploratory factor analysis (EFA) based on polychoric correlations was implemented in the R environment for statistical computing (R Core Team, 2015) using the Psych package (Revelle, 2015). Several indexes of fit were examined in concert to determine the best factor solution and the optimal number of factors for extraction including the root mean square error of approximation (RMSEA), root mean square of the residuals (RMSR), parallel analysis, and minimum average partial test (MAP; Velicer, 1976). Missing data ranged from 0% to 5% and did not meet the missing completely at random (MCAR) assumption. As a result, full information maximum likelihood estimation was applied for all correlational and EFA analyses. Factor scores were correlated with well-established measures of parenting, narcissistic traits, self-esteem, and entitlement to determine the construct validity of PISPS relative to the theorized nomological network. We applied bootstrapping (5,000 resamples) as a more robust analytic method because several of the variables did not meet the normality assumption. Correlational findings were the same with and without bootstrapping, and we chose to report bivariate correlations without bootstrapping for ease of interpretation. Finally, hierarchical regression analyses examined the incremental validity of the PISPS above and beyond parental warmth, rejection, and overcontrol.

Results and discussion

Item culling per classical test theory

Based on corrected item–total correlations, two items were removed: Item 3 (“I never got away with anything with my mother,” r = .019) and Item 19 (“My mother did not allow me to do anything before chores,” r = −.003). Items 23, 24, 25, and 29 were dropped due to low interitem correlations (r < .20; see Table 1). Items 5, 9, and 11 were also dropped based on high interitem correlations greater than .65. In sum, a total of nine items were dropped, reducing the overall scale to 26 items.

Ordinal exploratory factor analysis

The results of the EFA based on polychoric correlations supported a three-factor solution. An examination of the initial scree plot using the “elbow criterion” suggested that three factors adequately described the data (eigenvalues = 9.81, 5.17, 1.09, 0.67, 0.50; scree plot provided as supplemental material; see Figure S.1). Parallel analysis suggested that three factors should be extracted. Fit statistics were generated for all solutions above the “elbow” to further confirm the best fitting solution for the sample. The one-factor solution suggested inadequate fit to the data (RMSEA = 0.14, RMSR = 0.16) and accounted for 27% of the variance in PISPS. The two-factor solution adequately fit the data (RMSEA = 0.09, RMSR = 0.05) and accounted for 43% of the variance. However, the three-factor solution resulted in the best fit to the data (RMSEA = 0.08, RMSR = 0.04) and accounted for 46% of the variance in PISPS; the MAP index also suggested a three factor solution. Given these results, we extracted three, moderately interrelated factors (between-factor correlations ranging from .41–.54).

The first factor comprised 12 items that loaded distinctly onto a single factor, with item content best described as forms of praise (e.g., “My mother told me I was special”). The second factor comprised 10 items that loaded onto forms of indulgence (e.g., “My mother told me that my teachers did not understand how special I was”). The third factor comprised three items that loaded onto a single factor, and a fourth item (Item 34) that loaded onto Factors 1 and 3. This fourth item was placed on Factor 3 because the face validity of this item’s content corresponded best with the remaining items on Factor 3. These four items describe forms of parenting that instill the importance of gaining status (e.g., “My mother encouraged me to be the leader in all activities that I was involved in”; see Table 1).

Construct validity via correlations with other variables

We next examined the PISPS factors in relation to constructs believed to be relevant to its nomological network (see Table 2). Tests of dependent correlations were calculated to compare the correlations manifested between the factors and correlates. This specific test was used to account for the interdependence (or correlations) between the factors. The three factors of the PISPS were differentially related to previously established parenting constructs, including parental warmth, rejection, and overprotection. Praise was significantly and positively related to warmth, significantly and negatively correlated with parental rejection, and not associated with overprotection. In contrast, indulgence was not correlated with parental warmth but was significantly and positively correlated with parental rejection and overprotection. Similarly, status was significantly positively correlated with parental rejection and overprotection. All three factors were positively correlated with total trait narcissism as measured via the NPI and grandiose traits as measured via the PNI. The magnitude of the correlation with the NPI was larger for status. Correlations with praise and indulgence were not significantly different, and neither were correlations among all factors and grandiose traits on the PNI. Notably, only indulgence and status were associated with vulnerable narcissistic traits, and the magnitude of the correlations was largest for indulgence. All factors were associated with higher levels of entitlement. Praise was positively associated with self-esteem, and indulgence and status were negatively correlated with self-esteem.

Table 2.

Study 1 correlations between Praise, Indulgence, and Status Parenting Scale factors and parenting, narcissistic traits, entitlement, and self-esteem.

	Praise	Indulgence	Status
Maternal praise	—
Maternal indulgence	.41	—
Maternal status	.47	.54	—
Maternal warmth	.54a	−.05b	.02b
Maternal rejection	−.23a	.33b	.26b
Maternal overprotection	−.02a	.10b	.19c
Narcissism (NPI)	.22a	.21a	.35b
Grandiose narcissism (PNI)	.23	.17	.22
Vulnerable narcissism (PNI)	.01a	.24b	.11c
Entitlement	.18	.24	.25
Self-esteem	.22a	−.24a	−.08b

Note. NPI = Narcissistic Personality Inventory; PNI = Pathological Narcissism Inventory. Superscripts across rows are different from each other at p < .01 based on tests of dependent correlations (e.g., correlations with ^a in the same row are not statistically different from one another). Correlations ≥ are |.08|significant at p < .05.

In terms of incremental validity, PISPS factors (Step 2) predicted unique variance in narcissism, entitlement, and self-esteem, above and beyond parental warmth, rejection, and overprotection (Step 1), although it should be noted that some of the parenting constructs in Step 1 also remained significant in Step 2. See Table S.1 in the supplemental materials. Specifically, praise emerged as a unique predictor of grandiose narcissism on the PNI (in addition to rejection and PISPS status) and self-esteem. Indulgence and status were uniquely and positively related to narcissism (both grandiose and vulnerable) and entitlement. Indulgence was also uniquely and negatively related to self-esteem.

Taken together, results indicate a three-factor structure for the PISPS with differential links with correlates of interest. The praise factor appears to be associated with a mixture of positive and negative correlates such as positive parenting practices and self-esteem as well as narcissism and entitlement. In contrast, indulgence and status appear to be linked almost entirely with negative outcomes such as negative parenting practices (rejection and overprotection), narcissistic traits (both grandiose and vulnerable), and entitlement. Finally, depending on the outcome studied, PISPS factors accounted for unique variance in outcomes.

Study 2

In the second study, the PISPS was validated in a separate sample using IRT analyses with the goal of replicating the findings from Study 1, testing the validity of a father version of the scale, and examining links to internalizing problems.

Method

Participants

Participants were 464 undergraduate students between the ages of 18 and 37 (M_age = 19.58 years, SD_age = 1.77, 56% female, 81% White), collected from a large university in the southeastern United States who participated in exchange for course credit. Most participants labeled their maternal and paternal caregiver as their biological parents (99% and 96% for mothers and fathers, respectively). Maternal and paternal caregivers were most likely to be married (84% and 87% for maternal and paternal caregivers, respectively). Household income ranged from less than $20,000 (2%) to over $80,000 (50%) per year.

Procedure

Participants were asked to provide demographic information and to retrospectively report on their mothers’ and fathers’ engagement in these parenting practices as well as other parenting behaviors. Participants also reported on their own mental and emotional health, personality traits, and self-esteem. All study procedures were completed online via Qualtrics and were in accordance with the university’s institutional review board. Informed consent was obtained from all participants.

Measures

Praise, Indulgence, and Status Parenting Scale development (shortened 26-item version).

Participants completed the shortened, 26-item PISPS. The items were slightly modified to create a father version. Parents rated items using a 7-point-scale ranging from 1 (never) to 7 (always).” See Table 3 for internal consistency coefficients.

Table 3.

Study 2 factor loadings resulting from exploratory factor analysis across the two Praise, Indulgence, and Status Parenting Scale versions.

	Mother version			Father version
	Praise	Indulgence	Status	Praise	Indulgence	Status
Praise
1. My caregiver told me I was special	0.76	−0.05	0.03	0.89	0.10	0.10
4. My caregiver would call me pet names such as “Little Princess” or “Handsome Little Prince”	0.54	0.12	0.13	0.63	0.24	0.08
6. My caregiver told me I could be anything I wanted	0.83	−0.08	0.09	0.80	−0.04	0.16
7. My caregiver told me I was the most beautiful, handsome, or pretty child	0.76	0.13	0.12	0.74	0.31	0.09
8. My caregiver told me that I was smarter or brighter than other kids	0.65	0.17	0.13	0.65	0.19	0.29
12. My caregiver told me I would be successful no matter what	0.72	0.04	0.13	0.77	0.06	0.20
17. My caregiver told me I was a gift	0.69	0.25	0.16	0.73	0.26	0.35
20. My caregiver told me that I was a “Godsend”	0.47	0.44	0.31	0.51	0.38	0.41
27. My caregiver believed I had exceptional talents and abilities	0.63	0.16	0.22	0.67	0.18	0.42
28. My caregiver praised me for virtually everything I did	0.64	0.48	0.15	0.66	0.37	0.32
Indulgence
2. My caregiver thought it was funny when I broke the rules	0.00	0.50	0.14	0.09	0.60	0.04
10. I received special (better) treatment than my peers	0.44	0.45	0.14	0.43	0.54	0.24
13. My caregiver told me that my teachers did not understand how special I was	0.13	0.60	0.37	0.27	0.60	0.39
14. My caregiver made sure I got special privileges at school	0.09	0.67	0.33	0.08	0.74	0.39
15. My caregiver gave me most everything I wanted	0.40	0.51	0.03	0.45	0.52	0.03
16. My caregiver told me I did not have to do chores or other tasks because I was too precious	0.00	0.84	0.20	0.09	0.86	0.19
18. My caregiver let me rule the house	0.02	0.87	0.14	0.14	0.88	0.21
21. My caregiver intervened when others criticized me	0.38	0.51	0.24	0.40	0.40	0.32
22. My caregiver tolerated my temper tantrums	0.11	0.74	0.01	0.14	0.78	0.11
26. My caregiver sometimes put me on a pedestal	0.40	0.67	0.11	0.38	0.65	0.22
30. My caregiver let me make important family decisions at a young age	0.19	0.63	0.27	0.14	0.62	0.34
Status
31. My caregiver told me that I had unique talents and abilities that others did not have	0.41	0.36	0.39	0.48	0.25	0.62
32. My caregiver encouraged me to pursue stardom	0.32	0.31	0.68	0.24	0.34	0.74
33. My caregiver encouraged me to be in beauty pageants or other competitions	0.00	0.51	0.59	0.11	0.52	0.54
34. My caregiver encouraged me to be the leader in all activities that I was involved in	0.43	0.07	0.56	0.45	0.07	0.56
35. My caregiver often told me, “You are going to be famous”	0.22	0.46	0.70	0.31	0.45	0.68
Internal consistency (α)	α = .91 (total scale)			α = .93 (total scale)
	.87	.86	.80	.91	.88	.82

Note. Shaded items were dropped due to poor fit after the initial item–response theory partial credit models were run. Largest factor loading in bold.

Several of the measures included in Study 1 were also administered in Study 2, including the demographics questionnaire, EMBU–C (α = .67 and .65 for the mother- and father-report versions of the scale), NPI (α = .85 for the total score), PNI (α = .84 and .94 for grandiose and vulnerable trait narcissism scales, respectively), PES (α =.89), and RSES (α = .89).

Psychopathology symptoms.

Symptoms were assessed with the Brief Symptom Inventory (BSI; Derogatis, 1993), a 53-item self-report scale measuring psychological distress over a variety of mental health symptoms during the past 7 days. The anxiety (α = .83) and depression (α = .88) subscales were used in this study.

Data analytic plan

To confirm the structure of the scale, we applied an IRT approach to validation. IRT has been identified as a particularly appropriate approach to the development and validation of ordinal scales measuring latent constructs and overcomes many of the well-established limitations of classical test theory approaches to psychometrics, including confirmatory factor analysis and assumptions for continuous data (de Ayala, 2009). To appropriately apply an IRT model to the data, the items must reflect unidimensional constructs. Thus, when validating scales comprising multiple factors, each factor or singular dimension is submitted to an individual IRT model. To examine the reduced 26-item version of the scale in the new sample, we first sought to replicate the factor structure found in Study 1 in the service of confirming that each of the three factors extracted could be submitted to separate IRT models across both the mother and father versions of the scale. To that end, an ordinal EFA based on polychoric correlations was performed for each version of the scale following the same procedures described under Study 1.

For each factor extracted from the EFA, a partial credit model (PCM) with stepwise-elimination (described in more detail later) was fit to the data. The model produces an overall item location parameter, which models the distribution of items across the latent continuum, and a transition location parameter, which provides the same information for each response category on an item. The model does not assume that the distance (in terms of latent trait level) is homogeneous between every response category (e.g., there might be a larger difference between “often” and “always” compared to “never” and “seldom”). Item fit was examined via chi-square tests (items with significant chi-squares were eliminated and the PCM was refit to the remaining items) and mean-square infit and outfit statistics. Infit and outfit statistics < 2.0 suggest acceptable fit, and statistics ≤ 1.5 are ideal and indicate that the item is productive for measurement. Finally, a Martin–Löf likelihood-ratio (L-R) test was used to evaluate the fit of each PCM, also providing an indicator of measurement invariance. Factor scores were correlated with well-established measures of parenting, narcissistic traits, self-esteem, entitlement, anxiety, and depression. Finally, hierarchical regression analyses examined the incremental validity of the PISPS above and beyond parental warmth, rejection, and overcontrol.

Results and discussion

Response-category reduction

An initial examination of the item characteristic curves (ICCs; graphical representations of the probability of endorsing each response category as the latent trait level increases) and location transition parameters showed that the response categories on the 7-point scale were not ordered numerically along the latent trait. The ICCs indicated that some of the lower and upper middle categories were unproductive for measurement, as the probability of endorsing each category was the same or lower than the neighboring categories at all levels of the latent trait. We therefore changed the response format from a 7-point to 5-point scale across all scale versions. The response categories sometimes and occasionally were combined, and the response categories often and most of the time were combined. The 5-point scale was the most stable and best performing scale version, and all EFA and IRT results presented here are in reference to the 5-point version.

Ordinal exploratory factor analysis

The results of the EFA replicated the three-factor solution found in Study 1. Findings were consistent across mother and father reports. The three-factor solution accounted for 54% of the variance in PISPS responses for the mother version of the scale, with fit statistics indicating acceptable fit to the data (RMSEA = 0.10, RMSR = 0.05). For the father version, the three-factor solution accounted for 61% of the variance (RMSEA = 0.12, RMSR = 0.05). The Velicer MAP test reached a minimum value at the three-factor solution (compared to the one- and two-factor solutions) for both versions. All fit statistics fell within the acceptable range with the exception of the RMSEA value (RMSEA > 0.1); all remaining indexes including the scree plot, MAP values, and RMSR for both versions of the scale indicated that the three-factor solution fit the data best.

An examination of the factor loadings supported the retention of the three dimensions derived in Study 1 (see Table 3). The first factor comprised 11 items that loaded distinctly onto the factor best described as forms of praise. These were largely unchanged from Study 1 to Study 2, with the exception of Item 10, which loaded on praise in Study 1, but loaded on the indulgence factor in Study 2. We kept Item 10 in the indulgence factor. The second factor comprised 10 items that loaded distinctly onto forms of indulgence, and were the same as those in Study 1. The third factor comprised five items that loaded distinctly onto forms of parenting that instill status. The status factor in Study 1 contained four items, rather than the five identified in Study 2. We kept this fifth item (31) on the status factor because of face validity and because it loaded on this factor for both versions of the scale. Tucker’s coefficients of congruence between factors of different versions of the scale within Study 2 were computed, and all fell within the acceptable range (.96–.99). Tucker’s coefficients of congruence between the three factors in Study 1 versus Study 2 also fell within the acceptable range (.98–.99).

IRT partial credit model and item fit

A PCM was fit to each of the three factors and for the mother and father versions of the scale, resulting in six PCMs. After stepwise elimination of misfitting items, 21 overlapping items remained that demonstrated good fit across scale versions (see Table 3). See the supplemental materials online for the infit and outfit statistics for the three factors for each scale version.

Construct validity via correlations with other variables

Correlations examined relations between the three PISPS factors and measures of parenting, narcissistic traits, entitlement, self-esteem, and internalizing symptoms (see Table 4). Tests of dependent correlations examined significant differences between correlations. Consistent with Study 1, the praise factor was positively associated with parental warmth and negatively associated with rejection for both mothers and fathers. Also similar was the absence of a link with parental overprotection. Greater levels of maternal and paternal praise were positively associated with grandiose narcissistic traits, entitlement (fathers only), and self-esteem. Praise was not related to anxiety or depression.

Table 4.

Study 2 correlations between Praise, Indulgence, and Status Parenting Scale factors and parenting, narcissistic traits, entitlement, self-esteem, anxiety, and depression.

	Mother version (n = 464)				Father version (n = 464)
	Praise	Indulgence	Status	Total	Praise	Indulgence	Status	Total
Parental praise	—			.81	—			.86
Parental indulgence	.43	—		.83	.51	—		.83
Parental status	.49	.59	—	.80	.59	.62	—	.83
Parental warmth	.30a	−.07b	−.03b	.10	.45a	.10b	.18b	.31
Parental rejection	−.20a	.09b	.08b	−.03	−.22a	.05b	.05b	−.07
Parental overprotection	.02a	.16b	.10b	.12	.00a	.10b	.10b	.07
Narcissism (NPI)	.10a	.30b	.33b	.28	.13a	.28b	.28b	.26
Grandiose narcissism (PNI)	.15	.18	.24	.23	.14	.16	.19	.18
Vulnerable narcissism (PNI)	.02a	.27b	.19b	.20	.04a	.27b	.15c	.16
Entitlement	.08a	.33b	.27b	.28	.12a	.33b	.23c	.26
Self-esteem	.21a	−.13b	.01	.03	.18a	−.15b	.04c	.04
Anxiety	−.01a	.36b	.18c	.22	.06a	.36b	.17c	.22
Depression	−.05a	.25b	.10c	.13	−.01a	.22b	.07a	.11

Note. NPI = Narcissistic Personality Inventory; PNI = Pathological Narcissism Inventory. Maternal and paternal version of the Egna Minnen Betraffande Uppfostran were used in correlations with the mother and father versions of the PISPS, respectively. Superscripts across rows within reporter = different from each other at p < .01 based on tests of dependent correlations (e.g., correlations with ^a in the same row are not statistically different from one another). Correlations ≥ |.10| are significant at p < .05.

The indulgence factor was associated with parental overprotection, rejection (by mothers), and warmth (by fathers). Maternal and paternal indulgence were positively associated with narcissistic traits (both grandiose and vulnerable), entitlement, and anxiety and depression. Both maternal and paternal indulgence were negatively correlated with self-esteem. The status factor was associated with paternal warmth, grandiose and vulnerable narcissistic traits, and entitlement for both maternal and paternal reports. Engagement in status by both mothers and fathers was associated with greater levels of anxiety, but not self-esteem or depression.

In terms of incremental validity, PISPS factors (Step 2) predicted unique variance in narcissism, entitlement, and self-esteem, above and beyond parental warmth, rejection, and overprotection (Step 1), although it should be noted that some of the parenting constructs in Step 1 also remained significant in Step 2. See Table S.2 in the supplemental materials. Specifically, praise emerged as a unique predictor of vulnerable narcissism (for mothers only) and self-esteem. Indulgence was uniquely and positively related to narcissism (NPI), vulnerable narcissism (PNI), and entitlement, and negatively related to self-esteem. Status predicted unique variance in total narcissistic traits on the NPI, grandiose traits on the PNI, and entitlement (mothers only).

Results of Study 2 offer additional support for the three-factor structure of the PISPS. As reviewed earlier, factors were differentially linked with correlates of interests for both mother and father versions of the scale. Patterns of correlations were largely consistent across mother and father versions of the scale. One notable exception was the correlation between paternal indulgence/status and paternal warmth. Similar to results from Study 1, praise was positively associated with self-esteem, indulgence was negatively associated with self-esteem, and status was not significantly correlated with self-esteem. Patterns of correlations in Study 1 and Study 2 offer generally consistent results. One discrepancy was the correlation between status and self-esteem in Study 1 and the absence of significant parallel correlations for mothers and fathers in Study 2, although it should be noted the correlation in Study 1 was quite small (r = −.08, p = .047). Similar to Study 1, PISPS factors accounted for unique variance in outcomes examined. In both studies, praise emerged as a unique predictor of self-esteem; indulgence uniquely predicted vulnerable narcissistic traits, entitlement, and self-esteem; and status uniquely predicted grandiose narcissistic traits and entitlement. Finally, results also extend Study 1 findings by demonstrating links between the PISPS factors and internalizing symptoms.

Study 3

The goal of Study 3 was to extend the research to a sample of parents reporting on their parenting and their children’s emotional and psychological functioning. We specifically examined relations to the Parental Overvaluation Scale by Brummelman et al. (2015) with the hypothesis that PISPS indulgence and status would be positively correlated with parental overvaluation. We also tested whether the PISPS can offer any unique parenting information above and beyond existing parenting scales.

Method

Participants

Participants were 321 mothers and 327 fathers (M_age = 35.79 years, SD_age = 6.81) of children age 8 to 12 (M_age = 10.16 years, SD_age = 1.22) recruited from Amazon Mechanical Turk. Parents self-identified as mostly as White (73%), Black (12%), or Latino (6%). Most parents were married (66%) or single (19%), and most reporters were the biological parent to the child (93%). Household income ranged from below $30,000 (15%) to over $80,000 (22%).

Procedure

Parents completed study questionnaires on the Qualtrics survey platform via Amazon Mechanical Turk and received $5 for their participation. To ascertain parent status, respondents had to pass a set of three screener questions (i.e., “What type of device are you using to access this survey?,” “Are you a parent?,” and “How old is your child/your children?”). All study procedures were completed online via Qualtrics and were in accordance with the university’s institutional review board.

Measures

PISPS.

The final 21-item PISPS from Study 2 was used with items modified for parents’ self-reporting of current parenting practices. Internal consistency coefficients were good for all three subscales and for mothers and fathers (α ranged from .85–.91).

Parenting.

Parenting was assessed using two measures—the Parental Overvaluation Scale (POS; Brummelman et al., 2015) and the Coping with Children’s Negative Emotions Scale (CCNES; Fabes, Eisenberg, & Bernzweig, 1990). The POS is a 7-item scale assessing parents’ overvaluation of their children (e.g., “My child is more special than other children”) with good reliability (α = .74). The CCNES presents parents with vignettes of hypothetical situations that would be distressing to children (e.g., “If my child is panicky and can’t go to sleep after watching a scary TV show, I would …”) and asks parents to rate the likelihood they would respond in supportive (e.g., “encourage my child to talk about what scared him/her”) and unsupportive ways (e.g., “tell my child he/she is over-reacting”). The CCNES dimensions yielded good internal consistencies (α = .94 and .93 for the supportive and unsupportive scales).

Child emotion regulation.

Emotion regulation was assessed via the Emotion Regulation Checklist (Shields & Cicchetti, 1997). This 24-item questionnaire asked parents to rate their child’s dysregulation (e.g., “Exhibits wide mood swings”) and emotion regulation abilities (e.g., “Can recover quickly from episodes of upset or distress”). Internal consistencies were good for both subscales (α = .85 and .72, respectively).

Child psychopathology symptoms.

Parents reported on child symptoms using the Brief Problem Monitor (BPM; Achenbach, McConaughy, Ivanova, & Rescorla, 2011), which yields internalizing, externalizing, and attention broadband scores as well as a total score. The total score (α = .91) was used in these analyses.

Results and discussion

As expected, the indulgence and status factors of the PISPS were positively correlated with the POS for both mothers and fathers. However, in contrast to hypotheses, praise was also correlated with parental overvaluation. Patterns of correlations between the PISPS factors and parenting and child variables were similar for mothers and fathers. The only significant difference was between the negative correlation between indulgence and supportive parenting for mothers and the parallel nonsignificant correlation for fathers. Patterns of correlations with child variables were largely consistent with the previous two studies such that praise was positively associated with child emotion regulation and not associated with symptoms (the same as Study 2). In contrast, indulgence and status were negatively correlated with child emotion regulation and positively correlated with emotion dysregulation and symptoms of psychopathology (the same as Study 2; see Table 5).

Table 5.

Study 3 correlations between Praise, Indulgence, and Status Parenting Scale factors and parenting, child emotion regulation, and child symptoms.

	1	2	3	4	5	6	7	8	9
1. Praise	—	.51	.65	.48	.25	.00	.18	−.08	−.06
2. Indulgence	.42	—	.76	.30	−.02	.52	−.29	.34	.34
3. Status	.51	.75	—	.34	.02	.39	−.12	.19	.23
4. Parental overvaluation scale	.48	.37	.41	—	.04	.15	.02	.08	.15
5. Supportive parenting	.17	−.24	−.12	.05	—	−.34	.41	−.27	−.23
6. Unsupportive parenting	−.04	.46	.32	.11	−.50	—	−.55	.57	.54
7. Child emotion regulation	.13	−.39	−.21	.06	.46	−.51	—	−60	−.50
8. Child emotion dysregulation	−.05	.39	.25	.02	−.35	.55	−.65	—	.72
9. Child symptoms	−.05	.29	.17	.05	−.22	.50	−.50	.77	—

Note. Mothers are below the diagonal and fathers are above the diagonal.

Correlations ≥ |.11| are significant at p < .05.

To test whether the PISPS can add value above and beyond other parenting scales, a hierarchical regression was conducted predicting child emotion (dys)regulation and psychopathology symptoms with the POS in Step 1, the supportive and unsupportive parenting variables in Step 2, and the three PISPS factors in Step 3. See Table S.3 in the supplemental materials. In Step 3, PISPS praise and indulgence accounted for significant variance in child emotion regulation and dysregulation. Praise (fathers only) and indulgence accounted for unique variance in child symptoms. These results indicate that the PISPS, and specifically the praise and indulgence factors, contributes to unique variance in child symptoms, suggesting that this measure could offer useful information that is not covered by other measures.

General discussion

The overarching aim of this study was to develop and validate a parenting scale to measure parenting practices consisting of praise, indulgence, and encouragement of children to pursue extrinsic, status-oriented goals (i.e., status). It was hypothesized, through a symbolic interactionism framework, that these parental practices would be critical to children’s development of the self and relevant outcomes such as self-esteem, narcissism, entitlement, regulation, and psychological symptoms. We developed and validated the PISPS using classical test theory (Study 1) and then further refined the scale in a second sample using IRT (Study 2). Both studies provided support for the PISPS and its three-factor structure, and factors were differentially related to outcomes in our samples of emerging adults. Finally, in Study 3, findings were extended to a sample of parents reporting on their own parenting and their child’s emotion regulation and symptoms of psychopathology. Overall, findings support the PISPS, its psychometric properties, and its unique contribution to youth outcomes.

Regarding the structure of the PISPS, three factors emerged—praise, indulgence, and status—and these three factors were supported by both classical test theory and IRT approaches in all three samples. The three-factor solution explained 46% of the variance, which is consistent with other parenting scales (e.g., Brummelman et al., 2015). Internal consistencies for each of the factors were good across studies. Across the three studies, PISPS factors were differentially linked with other parenting dimensions, narcissistic traits, entitlement, self-esteem, emotion dysregulation, and psychological symptoms. Finally, we tested our hypothesis that the PISPS could offer useful parenting-related information that was not otherwise captured by established parenting scales. Our findings indicate that, depending on the outcome studied, the PISPS factors predicted unique variance in outcomes above and beyond parental warmth, rejection, and overprotection (Studies 1 and 2) and parental overvaluation, supportive, and unsupportive parenting (Study 3). These findings suggest that the PISPS has good potential to be useful in our understanding and measurement of parenting associated with a broad spectrum of outcomes related to adaptive self-esteem, inflated self-esteem, narcissism, entitlement, and psychopathology. Our findings suggest that appropriate praise appears to be well matched to other positive parenting strategies and associated with positive outcomes. However, concerns arise when parents indulge children or when children are encouraged to seek status over other individuals. Next, we elaborate on findings related to each factor, including validity evidence across studies.

The praise factor of the PISPS comprises communications that convey to children that they are unique and special and that offer encouragement toward future endeavors (praise; e.g., “My caregiver told me I could be anything I wanted”). Consistent with hypotheses, parental praise was consistently associated with supportive parenting, including parental warmth and low parental rejection. Not previously hypothesized was a relatively strong correlation between parental overvaluation and praise. This was unexpected given previous findings linking overvaluation with poor outcomes (Brummelman et al., 2015) and our current findings that praise was associated with positive outcomes. This suggests that parental overvaluation might be necessary but not sufficient to lead to the actual parenting practices that convey messages of overvaluation. For example, parents might hold the belief that their child is superior to other children, but they might not act on those beliefs. Importantly, the magnitudes of the correlations with other parenting constructs indicate that this praise factor is not simply a reflection of those already-established constructs. Further supporting the incremental validity of this factor were the results of the hierarchical regression showing that praise predicted unique variance in self-esteem. These results are consistent with previous findings showing that parental warmth and support are associated with self-esteem (Murray et al., 2000). Furthermore, praise was related to better emotion regulation and unrelated to symptoms of psychopathology, suggesting it is a largely adaptive parenting approach associated with desired outcomes in young people.

Two other parenting components emerged that center on excessively praising the child and encouraging the child to strive for extrinsic, status-oriented goals—indulgence and status. These factors differed from praise in that they were associated with generally negative parenting practices and outcomes. Maternal indulgence and status were not correlated with parental warmth (in Studies 1 and 2), yet we noted that when looking at fathers in Study 2, these links were significant and positive. This is a notable parent gender difference suggesting that fathers who overpraise and encourage the pursuit of extrinsic goals are also viewed as exhibiting warmth. If interpreted within the context of gender roles in parenting, it could be that positive involvement and engagement by fathers, regardless of the specific content, is viewed as particularly salient by children (e.g., Thomassin & Suveg, 2014). It would be worthwhile to examine these gender differences in future research also within the context of child gender. One discrepant finding emerged between Studies 1 and 2, which is that maternal indulgence and status were positively associated with maternal rejection in Study 1, and these correlations were much smaller in Study 2 (and nonsignificant for maternal status and for fathers). This might be due to the shortened scale used in Study 2 (vs. Study 1). The similar correlations between mothers and fathers in Study 2, which used the revised PISPS, lead us to believe that these relations are nonsignificant.

Despite very similar patterns of correlations between indulgence and status, the two factors differed primarily in their relation to narcissistic traits and self-esteem. Whereas indulgence uniquely predicted vulnerable narcissistic traits and self-esteem, status uniquely predicted grandiose narcissistic traits but not self-esteem. When positive parental messages are contingent on children achieving status, the contingent nature of these messages can create a sense of insecurity, ultimately affecting children’s self-esteem. Our findings support recent conceptualizations of narcissism and self-esteem as distinct constructs (see Brummelman et al., 2016) with positive correlations between indulgence and narcissism and negative relations with self-esteem. Indulgent parental practices are likely problematic because they communicate to the youth that they are superior, that others should accommodate them, and that they should dominate others. When faced with feedback that is incompatible with these messages (e.g., not making the team or poor academic performance), it might be difficult for youth to reconcile this conflict. Interestingly, Wuyts, Vansteenkiste, Soenens, and Assor (2015) found that parents themselves might be reacting to social messages and pressures to engage in praise, indulgence, and status behaviors. They found that parents were more likely to use an achievement-oriented motivation strategy for their children when they based their own self-worth on the successes of their children. This, in turn, was related to perceptions of social pressure to be an achievement-promoting parent. As such, it is important to recognize praise, indulgence, and status behaviors both on the individual level (i.e., as something parents differ on individually) as well as part of a larger social pressure to adopt such practices. It should also be noted that status did not uniquely predict child emotion regulation and symptoms in Study 3. It could be that status parenting during middle childhood does not yet correlate with child outcomes. Further research is needed to clarify the distinct constructs associated with status parenting at different developmental stages.

Limitations and future directions

Several limitations suggest avenues for future research. Studies were cross-sectional and correlational. It is thus not possible to draw causal conclusions about early childhood parenting communications and outcomes in young people. This study did not examine the temporal stability of the PISPS or the test–retest reliability. It is also possible that shared method variance accounted for some of the links between constructs because validation was based on self-report measures. In Studies 1 and 2, parenting reports were retrospective and participants were instructed to think about parenting when they were “young and growing up,” which could have affected the accuracy of the reports. With that said, in Study 3, we tested the PISPS in a sample of parents reporting on their current parenting practices and found similar results. The samples were primarily White; it is thus unclear the extent to which current findings would generalize to other samples, particularly more diverse samples, given the link between culture and parenting (e.g., Morelen, Jacob, Suveg, Jones, & Thomassin, 2013). The PISPS assesses a unique constellation of parenting practices, and future research would benefit from examining praise, indulgence, and status related parenting and other parenting constructs in a more nuanced way such as by observing parent–child interactions. Further examination of praise, indulgence, and status parenting should also address important nuances related to gender interactions between parent and child.

Implications

Despite study limitations, the consistency of findings across three different samples offers evidence for the robustness of the PISPS. Additional analyses comparing the PISPS against other established parenting measures highlight the PISPS’s potential to contribute in a significant way to our understanding and measurement of parenting. This investigation builds on the parenting, self-esteem, and narcissism literatures by contributing a questionnaire that has the potential to capture an important constellation of parenting practices that contribute to adaptive outcomes such as self-esteem and maladaptive outcomes such as narcissistic traits, entitlement, dysregulation, and psychological symptoms. To date, no questionnaire exists that assesses the full spectrum of parenting practices, some of which are beneficial for children and some of which might confer risk for poor child outcomes.

A more nuanced understanding of praise, indulgence, and status parenting could also have important clinical implications. For example, the etiology of narcissistic personality disorder is complex, but previous research has highlighted parenting as an important contributing factor (Kernberg, 1975; Millon, 1981). Identifying the complex relations between praise, indulgence, status, and views of the self (e.g., self-esteem, narcissism) might provide a means to prevent or reduce the effect of these parenting approaches on negative outcomes. Further, several evidence-based therapies for childhood problems place emphasis on parental praise, particularly as a means of positively reinforcing good behavior. Practitioners and clients alike might benefit from a deeper understanding of the distinction between praise (and the mechanisms by which praise could enhance self-esteem) and overpraise, including how overpraise, or indulgence, might paradoxically confer risk for unstable, inflated self-esteem, entitlement, and narcissism.